26 F.(2d) 126 established by direct testimony, the corporation had no 1917 earnings with which to pay the dividends. It appearing that the dividends were actually paid from the 1916 earnings, the Supreme Court held that the tax rate for that year applied, reversing the Circuit Court of Appeals, which had erroneously held that, if the dividends were paid in 1917, the earnings for the whole of that year should be found, and, in case such earnings were sufficient to pay all the dividends paid within the year, the dividends must be deemed to have been paid from them. Touching upon the main question which counsel sought to have the court decide-as to whether the rate of taxation was determined by the declaration or payment of the dividend, in event of sufficient current earnings to pay the tax-the court, by Mr. Justice Brandeis, said: "Since two of the dividends paid in 1917 were declared in 1916, it becomes necessary for us to consider whether these also are to be deemed distributions made in 1917, as it is only to such that the section applies. It declares that the dividend is income of the shareholders in the year in which it is 'received.' We think it clear that, for this purpose, the date of payment, not the date of the declaration of the dividend, is the date of distribution; and, as all the dividends here in question were paid in 1917, the provision as to the rate is applicable to all. As there were no earnings in 1917 prior to the dates of the payments, and as there were confessedly ample accumulated earnings of 1916 prior to the declaration of the several dividends, we have no occasion to consider other questions which were argued. The judgment of the Circuit Court of Appeals is reversed; that of the District Court is affirmed." Shortly after the opinion in Mason v. Routzahn, supra, was handed down, the Circuit Court of Appeals for this Circuit de cided United States v. Phillips, No. 3485, 24 F. (2d) 195, October term, 1927. The facts of that case were very similar to those in Mason v. Routzahn. The District Court (12 F.[2d] 598) had held that the rate of taxation was determined by the time of the declaration, not the payment, of the dividend, and had found for the plaintiff. The Court of Appeals reversed this judgment, citing Mason v. Routzahn as authority for the proposition that "distribution" (in that part of section 31(b) of the Revenue Act of 1917 which declares that "distribution" made to stockholders "in the year 1917 shall be deemed to have been made from most recently accumulated undivided profits or surplus") indicates the time of re26 F. (2d)-9 ceipt by the stockholders, and not the time of the declaration of the dividend by the directors. In passing upon the issues of United States v. Phillips, the Court of Appeals, by Judge Woolley, said: "The question on the second ground for increasing the assessment had its rise in opposite meanings which the parties attributed to the words of subsections (a) and (b) of section 31 of the Revenue Act of 1916, as amended by the Revenue Act of 1917 (40 Stat. 338, 377 [Comp. St. § 6336z]), the first of which declares the term 'dividends' to mean any 'distribution' made or ordered by a corporation out of its earnings or profits, and the second provides that any 'distribution' made to stockholders 'in the year 1917 * shall be deemed to have been made from the most recently accumulated undivided profits or surplus, and shall constitute a part of the annual income of the distributee for the year in which received.' The question, regarded by the trial court as one purely of law, was whether distribution should be considered made when the dividend is declared, as contended by the plaintiff, or when the dividend is paid, as contended by the government. The learned trial court held that distribution occurs when the dividend is declared. That question is now ruled adversely to the plaintiff by the decision of the Supreme Court in Mason v. Routzahn, Collector [275 U. S. 175], 48 S. Ct. 50, 72 L. Ed., following a related decision in Edwards, Collector v. Douglas, 269 U. S. 204, 46 S. Ct. 85, 70 L. Ed. 235, wherein that court held that 'the date of payment, not the date of the declaration of the dividend, is the date of distribution."" From the foregoing it is apparent that it would be our duty, were we passing upon the meaning of the word "distribution" in section 31(b), as determining the rate of taxation, to hold that it refers to the time of the receipt of the dividend by the stockholders rather than to the declaration of it. In holding that the word, as used in the exemption proviso attached to section 31(b), has a different meaning, we trust it will be distinctly understood that we are not undertaking to disregard the authority of the Court of Appeals or to question the accuracy of its decision. As it seems to us, the legislative history of the proviso makes it plain that Congress, no matter what the meaning given to the word "distribution" in the other part of the section, had the declaration of the dividend in mind in its use in the exemption clause. Corporation profits accumulated prior to March 1, 1913, are not subject to income tax. In construing the first Income Tax Act (38 Stat. 114), the Circuit Court of Appeals for the Second Circuit (United States v. Guinzburg, 278 F. 363) held a dividend declared before March 1, 1913, payable on July 1, 1913, to be a capital account and so free of income tax in the hands of a stockholder. Shortly before that decision, however, came Lynch v. Hornby, 247 U. S. 339, 38 S. Ct. 543, 62 L. Ed. 1149, which held a dividend from profits accumulated prior to March 1, 1913, declared and paid after that date, to be taxable income of the stockholder. Congress, without doubt having this case in mind, to protect dividends from accumulations antedating March 1, 1913, from such attack as was successful in Lynch v. Hornby, enacted section 2(a) of the Revenue Act of 1916 (Comp. St. § 6336b [a]), which permitted corporations to declare dividends from March 1, 1913, accumulations tax free to the stockholder. Section 2(a) of the Act of 1916 is as follows: "That the term 'dividends,' as used in this title shall be held to mean any distribution made or ordered to be made by a corporation, joint-stock company, association, or insurance company, out of its earnings or profits accrued since March first, nineteen hundred and thirteen." Section 2(a) of the Act of 1916 (Comp. St. § 6336b (a) was re enacted in section 31(a), as added by section 1211 of the War Revenue Act of 1917; Comp. St. § 6336z (a). Subsequent to its first passage, and prior to the proposal of section 31(b) as a law, corporations declared dividends upon March 1, 1913, surplus which were free of tax. This immunity from tax led to certain difficulties. Dividends were declared from profits accumulated prior to March 1, 1913, not from large current earnings, and the government secured no taxes upon them. To meet this condition, section 31(b), without the exemption proviso in question in the instant case, was conceived. It first came to public notice when reported to the Senate by Senator Simmons, chairman of the Senate Finance Committee. With the proposed Revenue Act was a report which called attention to the changes proposed, among them the subject-matter of section 31(b). When the proposed section was called to public attention, many protests were filed with the Conference Committee. The chief source of complaint was the retroactive feature of the proposed act. Many corporations, acting under section 2(a) of the Act of 1916, with the acquiescence of the Bureau of Internal Revenue, had declared dividends in 1917 from March 1, 1913, surplus. The proposed law was retroactive, and would have affected many dividends declared subsequent to January 1, 1917. Those dividends, instead of being tax free, as thought at the time they were declared, would be subject not only to tax, but to a much larger tax than had theretofore been laid. In view of the surprise feature of the proposed enactment, and the fact that many stockholders relying upon existing law would be subject to a tax, wholly unexpected and unprecedentedly large, upon dividends which might already have been expended, it was felt that the proposed change was exceedingly unfair, and possibly could not be sustained in law. Congress gave heed to the protests against this retroactive feature, and, when the bill came from the Conference Committee, it contained section 31(b) in its present form, with its exemption of all distributions prior to August 6, 1917. In discussing the act on the day of its passage by the Senate, Senator Simmons (Congressional Record, Sixty-Fifth Congress, p. 7615) said: "Under existing law and in the House bill the accumulated surplus of corporations when distributed, whether in cash or in stock, become subject to surtax of the stockholder receiving it unless the surplus thus distributed was accumulated prior to March 1, 1916 (1913), the date of the adoption of the income tax amendment to the Constitution. Surplus acquired before that date is not and cannot be subject to an income tax of any kind. In this condition of the law it was clearly to the interest of the corporation to distribute its surplus accumulated before 1913 instead of that more recently accumulated." "In order to be just to the corporations and at the same time to encourage them to distribute their annual earnings, your committee reported to the Senate and the Senate adopted an amendment providing that the surplus when distributed shall be taxed to the distribution (distributee) at the rate prescribed by law for the year in which they were earned, but also provided that any distribution of surplus made in 1917 or subsequent tax years should be deemed to have been made from the most recently accumulated undivided profits, thereby making it necessary for the corporation to distribute its surplus earned since March 1, 1913, before that earned prior to that time." "The conferees agreed to these amendments, but added an amendment to the effect that the later amendment—that is to say, the one requiring the distribution to be made 26 F.(2d) 131 from the most recently acquired surplus should not apply to any distribution made prior to August 1 (6), 1917, out of earnings accrued prior to March 1, 1913." The reason for exemption of distributions of March 1, 1913, profits, made prior to August 6, 1917, is plain. August 6, 1917, was the date upon which section 31 (b) was called to public attention by report of the Finance Committee of the Senate. If corporations, after such notice of the proposed enactment, declared 1913 dividends, they did so with knowledge of the possibility that such dividends would be subject to the high 1917 tax rate. Before that date no notice of the section had been given, and all dividends theretofore declared in good faith were entitled to the protection of Congress. In view of the circumstances of, and the reasons for, the addition of the proviso by which distributions of March 1, 1913, accumulations were exempted from taxation if made prior to August 6, 1917, it is inconceivable that Congress intended "to be just to" some corporations which had been misled by its previous actions, and to allow others, in practically the same situation, to suffer the effects of a retroactory law. Such an intention would have to be presumed if the word "distribution" in the exemption is construed to mean the payment of the dividend to the stockholders rather than the declaration of it. In the instant case the interpretation would mean the imposition of a heavy tax upon a dividend which had been declared by a corporation in the well-founded belief that it was free of tax. The declaration of the dividend bound the corporation, which could revoke it for no reason short of fraud. It is plain, we think, in view of the purpose of the addition of the proviso to the section, that Congress did not intend the inequality of treatment attributed to it, but did intend to protect all dividends declared in good faith before notice had been given of any proposed change of law. In arriving at the foregoing conclusion, we have not ignored the general rule of construction that, when a word has been used several times, and in different connections, in a statute, the same meaning of it is to be presumed throughout. The present statute, it would seem, furnishes an example of a well-defined exception to the rule. The rule is of great force when the word to be interpreted is found in a statute prepared by one person and all parts thereof joined together at about the same time. But that is not the case presented by the proviso to section 31(b). The proviso was not in the original section in any form. To remedy a bad feature of the original draft of the act, it was added while the bill was before the Conference Committee, and did not publicly appear until October 2, 1917, the day before the act was passed. This fact in itself weakens the presumption, and, when considered in connection with the evil sought to be remedied by the proviso, has tended to take away any difficulty in arriving at our conclusion that Congress, by the use of the word "distribution" in the proviso, intended the declaration of the dividend rather than its payment. This conclusion requires the entry of judgment in favor of the plaintiff. LOGAN-GREGG HARDWARE CO. v. HEIN- No. 3564. 1. Internal revenue 38(1)-Delivery of certificate of overpayment constituted a payment sufficient to authorize recovery against collector, for unlawful collection of taxes. Delivery to collector of internal revenue of certificate of overpayment of taxes constituted payment sufficient to authorize recovery from collector of any amount so paid, providing payment was made pursuant to collection of taxes not lawfully due. 2. Internal revenue ~7(3), 9(27)—Dividends pursuant to resolution of January 25, 1918, did not constitute part of "invested capital," though not actually withdrawn within first 60 days of taxable year (Revenue Act 1918, S$ 201(e), 326, Comp. St. §§ 6336b(e), 6336161). Under Revenue Act 1918, § 201(e), Comp. St. § 6336gb (e), declaring that any distribution made during first 60 days of taxable year ing of preceding taxable years, dividends declared pursuant to resolution of company, did not constitute a part of "invested capital," as defined by section 326 (Comp. St. § 63366i), for shall be deemed to have been made from earn the year 1918, though amount of dividends declared were not actually withdrawn in cash during first 60 days of taxable year, and withdrawal of invested capital was properly declared as of January 25, 1918, pursuant to treasury regulation 45, art. 858. [Ed. Note. For other definitions, see Words and Phrases, First and Second Series, Capital Invested.] 3. Statutes 219-Treasury Department regulation promulgated by authority of Congress, and impliedly sanctioned by re-enactment of law, is entitled to great consideration. Treasury Department regulation, promulgated by authority of Congress, is entitled to great consideration, particularly where Congress has several times re-enacted the provisions of law which regulations were designed to interpret, thereby impliedly sanctioning regula- note, in amount of $1,000, was paid; and on tion. At Law. Action by the Logan-Gregg Hardware Company against D. B. Heiner, Collector of Internal Revenue for the Twenty-Third District of Pennsylvania. Judgment for defendant. Smith, Shaw & McClay, of Pittsburgh, Pa., for plaintiff. John D. Meyer, U. S. Atty., of Pittsburgh, Pa., for defendant. GIBSON, District Judge. The instant action has been brought by the plaintiff to recover the sum of $5,209.10, with interest. The principal amount, it is claimed, was exacted from plaintiff by defendant in excess of the income and excess profits tax for the year 1918 that was actually due from plaintiff for that year. Finding of Facts. The Logan-Gregg Hardware Company is a corporation organized under the law of the state of Pennsylvania, and has maintained an office in the city of Pittsburgh, in the Western District of Pennsylvania and within the Twenty-Third internal revenue district of Pennsylvania. D. B. Heiner, since August 1, 1921, has been collector of internal revenue for the collection district above mentioned. On January 25, 1918, the board of directors of the plaintiff company duly passed the following resolution: "Resolved that we declare special salaries out of the earnings of the year 1917, representing 25 per cent. on the common stock and participating certificates, payable in cash or notes of the company bearing 5 per cent. interest, at the convenience of the company." On January 31, 1918, in pursuance of the preceding resolution, plaintiff paid to a stockholder the sum of $6,345.70 in cash, and canceled his debt to the company in amount of $1,654.30; and at the same time it issued notes bearing 5 per cent. interest per annum, to other stockholders, in the aggregate amount of $60,325; and on April 17, 1918, pursuant to the same resolution, it issued like notes, in the aggregate amount of $21,000, to the remaining stockholder, who could have received such notes on January 31, 1918, but who was absent in Europe at that time and did not demand his notes until April 17, 1918. Said notes were delivered in payment of dividends. On February 25, 1918, one of such notes, in amount of $2,000, was paid to a stockholder by the company; on May 28, 1918, another May 31, 1918, others of such notes, in the aggregate amount of $70,825, were paid; and the last payment on said notes was made on November 16, 1918. At the time said dividend was declared, and said notes were delivered on January 31, did not have on hand actual cash sufficient to 1918, and thereafter the plaintiff company pay the dividend declared in full, and issued notes, as aforesaid, to maintain in the treasury cash sufficient to meet the spring purchases of the company. The plaintiff company, on June 27, 1918, paid in cash a dividend of $21,438, on the common stock of the company, and a stock dividend of $35,730, and during the same year paid dividends aggregating $11,987.50 on the preferred stock. These dividends were declared subsequent to the resolution of January 31, 1918. At the dates appointed by law, the plaintiff company filed with defendant's predecessor in office its income and excess profits tax return for the year 1918, upon which return income and excess profits tax was assessed against the plaintiff in the sum of $116,928.49. An amended return was subsequently filed, whereupon the Commissioner of Internal Revenue reduced the assessment to the sum of $112,343.61, of which amount $101,356.72 was paid to defendant's predecessor in office, and of the balance $2,100.16 was paid to defendant in cash on February 13, 1926, and on July 14, 1923, $8,886.73 was paid to him by the application of a certificate of overassessment in plaintiff's favor. In fixing plaintiff's income and excess profits tax for 1918, at the sum of $112,343.61, as stated, the Commissioner of Internal Revenue determined plaintiff's net income at the sum of $220,492.98, and its invested capital at $809,217.83. In arriving at the sum last mentioned as the amount of plaintiff's invested capital in 1918, the Commissioner deducted the total amount of the dividend declared on January 25, 1918, as aforesaid, to wit, $89,325, from the amount of plaintiff's invested capital for the year 1918, as claimed by plaintiff, prorating the amount of said dividend from January 25, 1918. Had the Commissioner not deducted the total amount of the dividend notes aforesaid, paid subsequent to March 1, 1918, from plaintiff's invested capital the average amount of such capital for 1918 would be $883,628.28, and its income and excess profits tax would have been $107,134.51. On or about September 23, 1925, the plaintiff filed with the defendant collector, in 26 F.(2d) 131 the form prescribed by the Commissioner of Internal Revenue, a claim for the refund of $5,209.10. This claim was based upon the amount of the said dividend declared by resolution of January 25, 1918, paid subsequent to February 25, 1918, deducted by the Commissioner from plaintiff's invested capital for the year 1918. It was rejected by the Commissioner on February 23, 1926. Opinion. The issue in the instant action springs from the determination of the amount of plaintiff's invested capital for the year 1918 by the Commissioner of Internal Revenue. The Commissioner held such invested capital to be $809,217.83. In arriving at this amount he deducted from plaintiff's capital the total amount of the dividend declared by the resolution of January 25, 1918, hereinbefore set forth. Plaintiff contends that the Commissioner's determination was erroneous, in that only $10,000 of the dividend had been paid within the first sixty days of the taxable year, and was deductible, and the balance, $79,325, had been actually paid at various times after that period, and when the current earnings of the year were sufficient to pay it—thus making a deduction of more than $10,000 from invested capital, contrary to law. The defendant, by enforcing the alleged erroneous deduction of the Commissioner, plaintiff claims, collected $5,209.10 more than was due from plaintiff in payment of its income and excess profits taxes for the year 1918, and such amount, with interest, plaintiff now seeks to recover. The defendant contends that the amount deducted, as aforesaid, from plaintiff's invested capital for the year 1918, was taken from it by the Commissioner in compliance with the governing statute and the lawful regulations established for its enforcement. He further urges, in defense of part of the claim, that this is a personal action, and that plaintiff under no circumstances can recover more than $2,100.16, the amount of cash actually turned over to defendant by it in payment of its 1918 taxes. [1] The second matter of defense alleged by defendant is not well founded, in our opinion. On July 14, 1923, plaintiff had a certificate of overpayment of taxes for the year 1917 to the amount of $8,886.73, and on that date, instead of cashing the certificate and turning the cash over to defendant in payment of its 1918 taxes, it transferred the certificate to him. The delivery of the credit to the defendant was a payment to him, and plaintiff is entitled to recover from defendant any amount so paid, provided the payment was made to defendant pursuant to the latter's collection of taxes not lawfully due from the plaintiff. [2] The conclusion just stated brings us to the consideration of the subject of the main issue in the case, the ruling of the Commissioner of Internal Revenue to the effect that all dividends paid pursuant to the resolution of January 25, 1918, reduced the plaintiff's "invested capital" as of that date by the amount of such dividends. "Invested capital" is defined by section 326 of the Revenue Act of 1918 (40 St. L. p. 1092 [Comp. St. § 63366i]). So much of that section as is material to the present matter is as follows: "Sec. 326 (a) That as used in this title the term 'invested capital' for any year means (except as provided in subdivisions (b) and (c) of this section): On April 17, 1919, under authority of the Revenue Act of 1918 (40 Stat. 1057) the Commissioner of Internal Revenue, with the approval of the Secretary of the Treasury, promulgated certain regulations for the enforcement of the act. Article 858 of such regulations (see Regulations 45, p. 187) was designed to clarify section 326 of the Revenue Act and to prescribe the practice of the Bureau of Internal Revenue thereunder. It is as follows: "Art. 858. Effect of ordinary dividend. —A dividend other than a stock dividend affects the computation of invested capital from the date when the dividend is payable and not from the date when it is declared, except that where no date is set for its payment the date when declared will be considered also the date when payable for the purpose of this article. For the purpose of computing invested capital a dividend paid after the expiration of the first sixty days of the tax |